Volatility Breakout Systems
by Linda Bradford Raschke
Breakout systems can actually be considered another form of swing trading, (which is a style of short term trading designed to capture the next immediate move). In other words, the trader is not concerned with any long term forecast or analysis, only the immediate price action.
Volatility breakout systems are based on the premise that if the market moves a certain percentage from a previous price level, the odds favor some continuation of the move. This continuation might only last one day, or go just a little bit beyond the original entry price, but this is still enough of a profit to play for. A trader must be satisfied with whatever the market is willing to give.
With a breakout system, a trade is always taken in the direction that the market is moving at the time. It is usually entered via a buy or sell stop. The bit of continuation that we are playing for is based on the principle that momentum tends to precede price. There is also another principle of price behavior that is at work to create trading opportunities. That is, the market tends to alternate between a period of equilibrium (balance between the supply and demand forces) and a state of disequilibrium. This imbalance between supply and demand causes “range expansion”, (the market seeking a new level), and this is what causes us to enter a trade.
There are several ways to create short-term volatility breakout systems. I have found that different types of systems based on range expansion test out quite similarly. Therefore, whichever method you choose should be a matter for your own personal preference.
In designing a system, one can choose to place an entry stop off either the opening price or the previous day’s closing price. This entry stop can be a function of the previous day’s range or a percentage of the previous 2.10-day range, etc. Mechanical exits can range from using a fixed objective level to using a time function such as the next day’s open or close. Most of these systems function best when a very wide stop is used.
Another way of trading the breakout mode is by using “channel breakouts” which is simply buying the highest high of the last seven days in the case of a 7-period channel or the highest high of the last 2 days in the case of a 2-period channel breakout. In the case of an inside day breakout pattern where one buys the high or sells the low of the previous bar, a 1-period channel breakout is actually being used for the trigger. The most famous long-term breakout system adapted by Richard Dennis for training the “Turtles” was the 4-week channel breakout originally designed by Richard Donchian. Other breakout systems can be based on chart patterns (i.e., Curtis Arnold’s Pattern Probability System), trendline breaks, breakouts above or below a band or envelope of prices, or variations of simple range expansion functions.
Training Benefits for the Novice Trader
Derived from Trading a Volatility Breakout System:
Trading a short-term breakout system can be one of the best exercises to improve your trading.
- First, it teaches you to do things that are hard to do – buying high or selling low in a fast moving market! For most people, this feels quite unnatural!
- Second, it always provides a defined money management stop once a trade is entered. Not adhering to a defined money management stop is the most common cause of failure among traders.
- Third, it teaches a trader the importance of follow-through once a trade is entered, as most breakout systems perform best when the trade is held overnight.
- Last, it provides a great means for traders to improve their execution skills. Most volatility breakout systems are fairly active compared to a long-term trend following system. A trader can gain skill in placing orders in a diverse number of markets. Having a mechanically defined entry point is sometimes just the thing needed to overcome a trader’s fear of pulling the trigger. The order is placed ahead of time and the market then automatically pulls the trader into a trade if the stop level is hit.
Even if a person prefers to ultimately enter orders using discretion, trading a mechanical volatility breakout system can still be an invaluable exercise. It should at least increase a trader’s awareness of certain types of price behavior in the marketplace, especially if one is conditioned to entering on counter-trend retracement patterns. It can’t but help impress upon one the power of a true trend day.
Pros and Cons of Trading a Breakout System:
Like most systems, volatility breakout systems will clean up in volatile or runaway markets but tend to thrash when conditions get choppy or volume dries up. I believe they are still among the most profitable type of system to trade, and I also feel they will continue to be profitable in the long run. They are “durable” and “robust”, though they tend to deteriorate when too large of an order is placed (i.e., greater than 50 contracts). However, so that you do not get the impression that there is a Holy Grail of systems, the following considerations should be kept in mind:
Entries can be nerve-racking, especially when the market is in a runaway mode. The best breakouts won’t give you retracements to enter on. You are either on board or you are not! If you conceptualize that the best breakouts turn into trend days, and are most likely to close on the high or low for the day, then it is not so difficult to enter. Usually it is best to have a buy/sell stop already resting in the marketplace.
Sometimes a market gaps open outside your initial entry level. These often turn into the best trades. They can also turn into the most aggravating whipsaws. Big gaps test out that one should still take the trade, but they will definitely add more volatility to your bottom line. If your trade gets stopped out and an new signal is given in the opposite direction, this reversing trade usually more than makes up for the first loss.
Whipsaws are a drag but they are also inevitable when trading a breakout system. Many times I have bought the highs and sold the lows. It takes a great deal of “confidence in the numbers” to trade this type of system. System testing should always be done for a minimum of 3 years, preferably 10. Be sure to then examine out of sample data to see how the system performed.
On balance, a volatility breakout system can be traded on most all markets. However, a market might be very profitable one year and yet perform mediocre at best the next. A portfolio of 10 to 12 markets seems to work well. The problem with trying to trade too many markets at once is that it can become quite difficult to keep up with the activity level if your parameters are fairly sensitive. Many times in systems development, people overlook what one person can realistically manage.
Enhancing a Basic Volatility Breakout System:
Adding filters can sometimes create further enhancements. Examples of types of filters include: indicators to determine whether or not a market is in a trending condition, seasonality, days of the week, or degree of volatility contraction already present in the market. Periods of low volatility in the market can be defined by a contraction in true range, a low ADX, or a statistical indicator such as a low historical volatility ratio or a low standard deviation.
A system then might look something like this:
- Initial volatility condition = true
- Buy or Sell on a stop based on the current bar’s open, plus or minus a percentage of the previous day’s range.
- Initial Risk management stops once a trade is entered.
- Exit strategy.
Types of variables which can be used in a simple range expansion breakout system:
- Period – is the breakout based on a function of the previous day or the previous 10-day period, for example?
- Range – does it use the average range for that period or the largest, smallest, or total range?
- Percentage – what percentage of the range is used? It is possible, for example, to use 120% of the previous 3-days’ total range.
- Base – is the range function added to the previous day’s close or the current day’s open. This function may also be added to the high or low of the previous bar or a previous period such as the last 10 days.
As a general rule of thumb, the greater the percentage factor used, the greater the percentage of winning trades will be. However, the overall system may be less profitable because fewer trades are taken.
Once again, an example of an initial condition might be: Enter a trade only on a day following the narrowest range of the last 7 days. Or, take a trade only if the market has made a new 20-day high or low within the last five trading days. Whenever you add a filter to a system, be sure to compare the results to a baseline and examine the difference in activity level.
- Time based (2nd day’s close, 1st day’s opening)
- First profitable opening (Larry Williams)
- Target or objective level (1 average true range, previous day’s high/low)
- Trailing stop (displaced moving average, parabolic, 2-day high/low)
Controllable Risk – the amount of risk which can be predetermined and defined by a money management stop.
Types of money management stops:
- fixed dollar amount
- function of average true range
- price level (i.e., bar high/low)
- Overnight exposure (close to open risk). You cannot exit a position when the market is not trading. Thus, you are subject to adverse gaps, which can be exaggerated by news or events.
- Slippage risk. Fast market conditions or thin, volatile markets often cause a trader to get filled at prices much worse than expected.
In general, the numbers behind most systems are very dependent upon capturing a few good trades. You can’t afford to miss the one good trade that can make your month.
Here are some tips for trading this or any other system:
- Gain confidence by first trading a system on paper.
- Make sure you can successfully trade a system mechanically before attempting to add any discretion.
- Track your actual performance against the mechanical system at the end of each day, rating your success by whether you can match the system’s performance.
- Monitor performance over an adequate sample, perhaps 100 trades or a set number of weeks. Do not let a down week or trade deter you.
- Manage the exits rather than filter the entries. It is impossible to tell in advance which trades will be the good ones. The one entry skipped might be the BIG ONE, and one can’t afford to miss it. Managing the exit means two things: The first, learn when it’s okay to let that occasional great trade run an extra hour or two before getting out; the second (which really depends on one’s skill level), learn to recognize a bit sooner when a trade is not working and exit just before the stop is hit.
All systems display subtle nuances and insights into the market’s behavior over time.
- Keep a notebook of your observations and patterns you notice. In this way you truly “make the system your own”.
- Never be concerned about how many other people are trading systems. If slippage seems excessive, it often suggests a significant breakout from a triangle or period of congestion. Remember: Something had to drive the market far enough to penetrate the breakout point in the first place!
If you are interested in reading more on the principle of range expansion/range contraction, Toby Crable pioneered some of the finest research in this area. I would strongly recommend his book Day Trading with Short Term Price Patterns and Opening Range Breakouts. The research in this book provides one of the most solid platforms for developing volatility breakout systems based off the opening price. Toby is a CTA who now manages over 100 million dollars based on some of these techniques. Another excellent value is Curtis Arnold’s book, PPS Trading System. He discloses a different type of system based on breakouts of traditional chart patterns as identified in Edwards and Magee’s book Technical Analysis of Stock Trends (another classic book which should be in every serious market technician’s library!). Curtis’s original system sold for over $2,000. The book is essentially the same thing and sells for less $50! These books, in addition to many others can be ordered through our bookstore.